Religious congregations reach many lives and play an important role in working for social change. Could they play a similarly powerful role in the fight against HIV/AIDS? Although there is some evidence for an affirmative answer, few studies have examined systematically the prevalence and predictors of congregational involvement in health programs in general, much less in HIV/AIDS services. The purposes of this research are to describe the processes by which HIV/AIDS prevention and care activities are implemented in congregations and to examine factors associated with such activities. It is informed by a conceptual framework in which congregations' capacity to engage in HIV/AIDS activities and their interest in doing so are affected by norms and attitudes (including attitudes towards HIV and homosexuality), the organizational structure of congregations and other community elements, various sets of resources (internal to the congregation and external), and demographics. It is hypothesized that some of these factors will predict a congregation's decisions to implement health programs of any kind; other factors will predict decisions to implement HIV/AIDS activities in particular, and a third set of factors will predict the type and intensity of activities implemented. Hypotheses will be further elaborated through site visits and interviews with pastoral and lay leaders at 12 congregations of varying denomination, ethnicity, and size in Los Angeles County, half of which have implemented HIV/AIDS programs and half of which have not but are otherwise comparable. Factors hypothesized to influence congregational involvement will then be tested quantitatively through a telephone survey of 800 congregations in ethnically diverse areas of the county disproportionately affected by HIV/AIDS. The effects of community characteristics will be examined by compiling neighborhood-level data on socioeconomic status, ethnic diversity, community need, and so forth. Hypotheses will be tested through hierarchical linear modeling techniques. Study findings will be of utility to county health departments and health care providers, which will be interested to know to what extent their efforts may be supplemented by congregations. Results will also be of interest to other organizations and agencies that seek to promote service provision in community-based settings and to train potential providers; knowledge of local barriers to, and facilitators of, HIV/AIDS engagement should help them target their efforts.